Adding set column values as min/max error bars on R

Question

Apologies if this has been asked before, but it is difficult to articulate in order to find an answer.

I have the following set of data below. In R, I would like to construct a bar plot for PV1 and PV2 for AvgRead and AvgUniq . For the error bars, I would like to set MinRead and MinUniq as the minima, and MaxRead and MaxUniq as the maxima.

If you could help that would be greatly appreciated. Again my apologies if this has been asked before.

         AvgRead  MinRead  MaxRead AvgUniq MinUniq MaxUniq
PV1          20     10        40     40       20     80
PV2          40     20        80     80       40     160

Answer 1

You need to reshape your data a little bit using the melt() and dcast() functions from reshape2 :

library(reshape2)
library(ggplot2)

df <- data.frame(
  row.names = c("PV1", "PV2"),
  AvgRead = c(20, 40),
  MinRead = c(10, 20),
  MaxRead = c(40, 80),
  AvgUniq = c(40, 80),
  MinUniq = c(20, 40),
  MaxUniq = c(70, 160)
)

df$name <- row.names(df)

df_molten <- melt(df)
df_molten$var1 <- substr(df_molten$variable, 1, 3)
df_molten$var2 <- substr(df_molten$variable, 4, 10000)

df_cast <- dcast(df_molten, name + var2 ~ var1, value.var = "value")

ggplot(data = df_cast, aes(x = name, y = Avg, fill = var2)) +
  geom_bar(stat = "identity", position = "dodge") +
  geom_errorbar(
    aes(ymin = Min, ymax = Max),
    width = 0.5,
    size = 1.3,
    position = position_dodge(0.9)
  )

EDIT: to change the order of the bars you need to change var2 to factors and sort the levels accordingly:

df_cast <- dcast(df_molten, name + var2 ~ var1, value.var = "value")
df_cast$var2 <- factor(df_cast$var2, levels = c("Uniq", "Read"))

ggplot(data = df_cast, aes(x = name, y = Avg, fill = var2)) +
  geom_bar(stat = "identity", position = "dodge") +
  geom_errorbar(
    aes(ymin = Min, ymax = Max),
    width = 0.5,
    size = 1.3,
    position = position_dodge(0.9)
  )